cs.CL（2025-08-07）

📊 共 33 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (22 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (9) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (22 篇)

#题目一句话要点标签🔗⭐
1 Do Machines Think Emotionally? Cognitive Appraisal Analysis of Large Language Models 提出CoRE基准，通过认知评估分析大语言模型的情感推理能力 large language model foundation model
2 ASCoT: An Adaptive Self-Correction Chain-of-Thought Method for Late-Stage Fragility in LLMs 提出ASCoT方法，解决大语言模型推理链中后期脆弱性问题 large language model chain-of-thought
3 Pruning Large Language Models by Identifying and Preserving Functional Networks 提出基于功能网络识别与保留的大语言模型剪枝方法 large language model ✅
4 Attention Basin: Why Contextual Position Matters in Large Language Models 揭示LLM注意力盆地现象，提出AttnRank重排序框架提升上下文学习效果 large language model
5 A Multi-Stage Large Language Model Framework for Extracting Suicide-Related Social Determinants of Health 提出多阶段大语言模型框架，用于抽取与自杀相关的健康社会决定因素 large language model
6 A Rose by Any Other Name Would Smell as Sweet: Categorical Homotopy Theory for Large Language Models 提出基于范畴同伦理论的大语言模型框架，解决语义等价语句概率分布不一致问题 large language model
7 LLMEval-Fair: A Large-Scale Longitudinal Study on Robust and Fair Evaluation of Large Language Models LLMEval-Fair：提出大规模动态评估框架，解决大语言模型评估中的数据污染和过拟合问题。 large language model
8 MultiCheck: Strengthening Web Trust with Unified Multimodal Fact Verification MultiCheck：通过统一的多模态事实核查增强Web信任 multimodal
9 Understanding and Mitigating Errors of LLM-Generated RTL Code 提出一种结合领域知识和调试循环的框架，显著提升LLM生成RTL代码的准确率。 large language model multimodal
10 FineDialFact: A benchmark for Fine-grained Dialogue Fact Verification 提出FineDialFact，用于细粒度对话事实核查的基准数据集。 large language model chain-of-thought
11 Efficient Reasoning for Large Reasoning Language Models via Certainty-Guided Reflection Suppression 提出CGRS方法，通过置信度引导抑制反思，提升大语言模型推理效率。 chain-of-thought
12 RTTC: Reward-Guided Collaborative Test-Time Compute 提出RTTC：一种奖励引导的自适应测试时计算框架，提升LLM在推理时的性能。 large language model
13 The World According to LLMs: How Geographic Origin Influences LLMs' Entity Deduction Capabilities Geo20Q+：揭示LLM在实体推断中受地理来源影响的偏见 large language model
14 LAG: Logic-Augmented Generation from a Cartesian Perspective 提出LAG：一种基于笛卡尔哲学的逻辑增强生成方法，提升知识密集型任务的准确性和减少幻觉。 large language model
15 MyCulture: Exploring Malaysia's Diverse Culture under Low-Resource Language Constraints MyCulture：提出马来西亚文化基准，评估低资源语言约束下LLM的文化理解能力。 large language model
16 Align, Don't Divide: Revisiting the LoRA Architecture in Multi-Task Learning Align-LoRA：通过对齐任务表征，提升LoRA在多任务学习中的性能 large language model ✅
17 Evaluation of Finetuned LLMs in AMR Parsing 通过微调LLM，在AMR解析任务上达到媲美复杂SOTA模型的性能。 large language model
18 NanoCodec: Towards High-Quality Ultra Fast Speech LLM Inference NanoCodec：面向高质量超快速语音LLM推理的低帧率音频编解码器 large language model
19 TASE: Token Awareness and Structured Evaluation for Multilingual Language Models TASE：多语言模型Token感知与结构化评估基准 large language model ✅
20 Decision-Making with Deliberation: Meta-reviewing as a Document-grounded Dialogue 提出对话代理以提升元评审效率 large language model ✅
21 Navigating Through Paper Flood: Advancing LLM-based Paper Evaluation through Domain-Aware Retrieval and Latent Reasoning PaperEval：通过领域感知检索和潜在推理，提升LLM在论文评估中的表现 large language model
22 BEE-RAG: Balanced Entropy Engineering for Retrieval-Augmented Generation BEE-RAG：通过平衡熵工程提升检索增强生成模型性能 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
23 Resource-Limited Joint Multimodal Sentiment Reasoning and Classification via Chain-of-Thought Enhancement and Distillation 提出MulCoT-RD以解决资源受限环境下的多模态情感推理与分类问题 distillation large language model multimodal
24 Cooper: Co-Optimizing Policy and Reward Models in Reinforcement Learning for Large Language Models 提出Cooper框架，通过协同优化策略模型和奖励模型，提升大型语言模型在强化学习中的推理能力。 reinforcement learning large language model
25 FedCoT: Communication-Efficient Federated Reasoning Enhancement for Large Language Models FedCoT：面向大语言模型的通信高效联邦推理增强框架 distillation large language model chain-of-thought
26 Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM 全面剖析大语言模型有害内容生成与安全防御技术 reinforcement learning RLHF large language model
27 MathSmith: Towards Extremely Hard Mathematical Reasoning by Forging Synthetic Problems with a Reinforced Policy MathSmith：通过强化策略合成高难度数学问题，提升LLM推理能力 reinforcement learning large language model chain-of-thought
28 Decoupling Understanding from Reasoning via Problem Space Mapping for Small-Scale Model Reasoning DURIT：通过问题空间映射解耦理解与推理，提升小规模模型推理能力 reinforcement learning distillation large language model
29 Learning to Reason for Factuality 提出结合精确性、细节和相关性的奖励函数，提升推理大模型的事实性。 reinforcement learning offline RL large language model
30 Do Political Opinions Transfer Between Western Languages? An Analysis of Unaligned and Aligned Multilingual LLMs 研究多语言LLM中政治观点的跨语言迁移现象，揭示文化政治倾向的潜在影响。 direct preference optimization large language model
31 CodeBoost: Boosting Code LLMs by Squeezing Knowledge from Code Snippets with RL CodeBoost：利用代码片段知识，通过强化学习提升代码大语言模型 reinforcement learning large language model

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
32 OmniEAR: Benchmarking Agent Reasoning in Embodied Tasks OmniEAR：用于评估具身智能体推理能力的综合性基准测试框架 spatial relationship embodied AI large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
33 How Do LLMs Persuade? Linear Probes Can Uncover Persuasion Dynamics in Multi-Turn Conversations 利用线性探针揭示LLM在多轮对话中的说服动态 manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页